As discussed in U.S. Pat. No. 6,340,004, the fuel economy of vehicles primarily depends on the efficiency of the mover that drives the vehicle. It is well recognized that the current generation of internal combustion (IC) engines lacks the efficiency needed to compete with fuel cells and other alternative vehicle movers. At least one study has recommended that auto manufacturers cease development of new IC engines, as they may be compared to steam engines—they are obsolete. The present invention is directed to an IC engine that is competitive with fuel cells in efficiency.
The following principles must be embodied in one engine in order for the engine to achieve maximum efficiency.
1) Variable Fuel Ratio and Flame Temperature
For ideal Carnot cycle efficiency:n=(Th−Tl)/Th
Where Th=highest temperature                Tl=lowest temperature (usually ambient tempature)        n=thermal efficiencyshows that the higher the temperature, Th, the higher the engine efficiency. This is not the case in real-world conditions. The basic cause of the breakdown in the Carnot cycle rule is due to the fact that the properties of air change as the temperature increases. In partcular, Cv, the constant volume specific heat, and Cp, the constant pressure specific heat, increase as the temperature increases. The ratio k, on the other hand, decreases with increasing temperature. To heat 1 lb of air at constant volume by 100 degrees F. requires 20 BTU at 1000 degrees F, but 22.7 BTU at 3000 degrees F. The extra 2.7 BTU is essentially wasted. At the same time, each increment of Th adds less and less to the overall efficiency. If Tl is 600 R, and Th is 1800 R (1340 degrees F.), n=0.66666. At Th=3600 (3140 degrees F.), n=0.83333, and at Th=5400 R (4940 degrees F.), n=0.88888. In the first instance, going from 1800 R to 3600 R netted an increase in n of 0.16666, whereas going from 3600 R to 5400 R netted only an increase in n of 0.0555, or ⅓ of the first increase. At the same time, the specific heat of air is a monotonic function of temperature, so at some point the efficiency gains from higher temperatures are offset by losses due to higher specific heats. This point is reached at around 4000 R.        
The most efficient diesels are large, low swirl DI (direct injection) turbocharged 2-strokes. These are low speed engines (<400 rpm) and typically have 100%–200% excess air.
The combustion temperature is proportional to the fuel ratio. A CI (compression ignition) engine will have a theoretical flame temperature of 3000–4000 R, as opposed to the SI (spark ignition) engine, which has a theoretical flame temperature of 5000 R. Note also that the reason the specific heat is increased is due to increased dissociation of the air molecules. This dissociation leads to increased exhaust pollution.
Ricardo increased the indicated efficiency of an SI engine by using hydrogen and reducing the fuel ratio to 0.5. The efficiency increased from 30% to 40%.
Hydrogen is the only fuel which can be used in this fashion. There are 2 basic types of ignition—spark and compression. This engine proposes to use hot air ignition (HAI), which allows variation in the fuel ratio similar to CI, but with the additional advantage that HAI does not require the engine do work to bring the air up to the temperature where it can be fired. All engines which claim to be efficient must use an ignition system which allows wide variations in the fuel ratio. An incidental advantage of this design is that because molecular dissociation is much less at lower temperatures, the resulting exhaust pollution (species such as nitrous oxide, ozone, etc) is also lessened.
2) Uniflow Design
Uniflow design, although it is more critical to a Rankine cycle engine, such as the Stumpf Unaflow steam engine, is also of importance to an IC engine. Generally speaking, in a uniflow design, the motion of the working fluid into and out of the cylinder does not cause degradation of the cycle efficiency. The uniflow design minimizes unwanted heat transfer between engine surfaces and the working fluid. Only two-stroke cycle IC engines can claim some kind of uniflow design.
Consider the typical four-stroke cycle Diesel engine:
1) Intake—Air picks up heat from the intake valve and from the hot head, piston and cylinder. Generally speaking, the air heats up from 100–200 F.
2) Compression—The air continues picking up heat, in addition to the work done on it by the engine.
3) Power—Air is hot after firing, and begins to lose heat to the walls. Luminosity of the diesel combustion process accounts for much of the heat lost. The short cycle time of a high speed Diesel engine holds these heat losses by conduction to a minimum.
4) Exhaust—During the blowdown, heat is transferred to the exhaust valve, and hence to the cylinder head.
The engine of the present invention has separate cylinders for intake/compression and for power/exhaust. The intake/compression cylinder is cool, and in fact during the intake and compression process, efforts can be made to create a nearly isothermal compression process by adding water droplets to the intake air. Addition of water droplets is optional and is not essential to the design, which has had its efficiency calculations performed without taking water droplet addition into account.
Addition of water droplets, of course, is impossible with a Diesel engine. A variation on this is used in SI engines, where the heat of vaporization of the fuel keeps the temperature down during compression. This is one reason why methanol, which has a high heat of vaporization, is used in some high performance engines.
The power/exhaust cylinder is the ‘hot’ cylinder, with typical head and piston temperatures in the range of 1000–1100 F. This necessitates the use of 18/8 (SAE 300 series) stainless steels for the head and piston, and superalloys for the valves. Any other suitable high temperature material, such as ceramics, can also be used in the application. Combustion temperatures are in the neighborhood of 2000–3000 F. The high heat of the combustion chamber prior to combustion reduces the heat transfer from the working fluid to the chamber during the power stroke. It also reduces the radiant heat transfer, however the larger reduction in radiant heat transfer comes from keeping the maximum temperature below 3000 F.
Thus, unwanted heat transfer is minimized in the engine of the present invention.
There are several dissociation reactions which become important absorbers of heat above 3000 F. The two most important are:2CO2≡2CO+O2  a)2H2O≡2H2+O2  b)
The production of CO, carbon monoxide, is particular undesirable, as it is a regulated pollutant. All of these reactions also reduce the engine efficiency.
3) Regenerator
In the use of a regenerator, the state of the art is not yet commercially feasible.
The principle of using a regenerator is not new. Siemens (1881) patented an engine design which was a forerunner of the engine of the present invention. It had a compressor, the air traveling from the compressor through the regenerator and into the combustion chamber. There are, however, some basic differences between the Siemens engine and the engine of the present invention:
1) Siemens proposed using the crankcase, rather than a separate cylinder, to compress the air. The engine appears to be a variation of Clerk's two-stroke cycle engine (1878).
The engine features are:    a) All of the compression occurs in the crankcase    b) Max compression occurs at the wrong time on the stroke. It should occur at piston TDC, not BDC. This is remedied by use of a reservoir. This greatly increases the compression work.    c) It is not clear that the Siemens engine can vary the fuel ratio. It is a spark ignition engine. Ignition is aided by adding oil to the regenerator as the fresh charge is passing through it.    d) The Siemens engine had the regenerator as part of the top of the cylinder head. The regenerator is exposed to the hot flame, and some burning occurs in the regenerator.
In the engine of the present invention, the compressor takes in a charge of air, compresses it and then transfers the entire charge through the regenerator. The compressed charge includes the space taken up by the regenerator. At TDC of the power piston, (60 deg. bTDC of the compressor) the valve opens and the charge flows from the compressor to the power cylinder. Near TDC of the compressor, fuel is sprayed into the power cylinder. Dead air is minimized throughout the system in order to realize the benefits of the regenerator and minimize compressor work. During combustion, the regenerator is separated from the burning gases by a valve.
Hirsch (U.S. Pat No. 155,087?) has two cylinders, passages between them, and a regenerator. Air from explosion in the hot cylinder is forced from the hot cylinder to the cold cylinder, where jets of water are used to cool the air and form a vacuum. It appears to be a hot air engine, does not specify an ignition system, and contains a pressure reservoir.
Koenig (U.S. Pat. No. 1,111,841) is similar in design to the engine of the present invention. It has a power cylinder and a compression cylinder and a regenerator in between. It does not specify the method of firing the power piston, and the valving is somewhat different. In particular, the inventor failed to specify a valve between the power piston and the regenerator. This results in the air charge being transferred from the compression cylinder into a regenerator at atmospheric pressure. As the compression cylinder is smaller than the engine cylinder, this will cause a loss of pressure during the transfer process.
Ferrera (U.S. Pat No. 1,523,341) discloses an engine with 2 cylinders and a common combustion chamber. It differs substantially from engine of the present invention.
Metten (U.S. Pat No. 1,579,332) discloses an engine with 2 cylinders and a combustion chamber between them.
Ferrenberg (see U.S. Pat Nos. 5,632,255, 5465702, 4,928,658, and 4,790,284) has developed several patents drawn to a movable thermal regenerator. The engine of the present invention has a fixed regenerator.
Clarke (U.S. Pat No. 5,540,191) proposed using cooling water in the compression stroke of an engine with a regenerator.
Thring (U.S. Pat No. 5,499,605) proposed using a regenerator in a gasoline engine. That invention differs greatly from present hot-air ignition system.
Paul (U.S. Pat Nos. 4,936,262 and 4,791,787) proposed to have a regenerator as a liner inside the cylinder.
Bruckner (U.S. Pat No. 4,781,155) has some similarities to the engine of the present invention. In this patent, fresh air is admitted to both the power cylinder and the compression (supercharger) cylinder. This differs from the engine of the present invention, as fresh air is only admitted to the compression cylinder. In addition, there is no valving controlling the flow of air through the regenerator. The cylinders are out of phase, but the phasing varies.
Webber (U.S. Pat No. 4,630,447) has a spark-ignition engine in which there are two cylinders out of phase with each other, with a regenerator in between. However, there is no valving controlling the movement of air in the regenerator as with the present invention.
Millman (U.S. Pat No. 4,280,468) has a single cylinder engine in which a regenerator is placed between the intake and exhaust valves on the cylinder head. Very different from the engine of the present invention.
Stockton (U.S. Pat No. 4,074,533) has a modified Sterling/Ericsson engine with intermittent internal combustion and a regenerator.
Cowans (U.S. Pat No. 4,004,421) has a semi-closed loop external combustion engine.
Several U.S. patents were mentioned in the above patents. The most common for the closely allied patents were: U.S. Pat Nos. 1,682,111, 1,751,385, 1,773,995, 1,904,816, 2,048,051, 2,058,705, 2,516,708, 2,897,801, 2,928,506, 3,842,808, 3,872,839, 4,026,114, 4,364,233, 5,050,570, 5,072,589, 5,085,179, 5,228,415
4) Low Friction & Compression Ratio
In a regenerative engine scheme, the compression ratio needs to be low. It turns out that having a low compression (and expansion) ratio has the following advantages:
1) low friction mean effective pressure (fmep). fmep consists of rubbing and accessory mep (ramep) and pumping mep (pmep). Because the engine of the present invention is not throttled, there is very little pmep. The pmep in the engine of the present invention will primarily come from transfer of the air from the compression to the power cylinder and is generally no more than 1–2 psi at 1800 rpm. Ramep should be very low, as peak pressures are low and compression ratios are low.
2) Efficiency is high. This is due to the fact that the waste heat is recovered from the exhaust. It is more efficient to have a low compression ratio and recover much waste heat than it is to have a high compression ratio and recover a small amount of waste heat. The low compression ratio engine acts much more like a Sterling engine and hence its maximum possible efficiency is greater.
Almost by definition, a high friction engine cannot be efficient. None of the engines wth regenerators in the patents mentioned having a low compression ratio, except Webber (U.S. Pat No. 4,630,447), which has a 4:1 compression ratio. Webber also calls his engine an “open cycle Sterling engine.”
The current state of the art as commercially practiced does not produce engines that have adequate fuel economy. The state of the art as practiced in the patent literature does not adequate regulate the air flow through the regenerator. For example, in Webber's patent, hot gases can transfer unimpeded from the hot side to the cool side after firing. As these hot gases are expanding, the reduction in volume in this movement causes loss of power and efficiency. The regenerator picks up combustion heat, not exhaust heat.